Reduce downtime for patching

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Reduce downtime for patching

satish
Dear erman,

We are on R12.2.5 and Ad/ txk delta 7 ATG_PF 5.

DB version 12.1.0.2

We are planning to enable java web start and for this the minimum ad/txk delta level is 8 and atg_ pf is 6.

1)Apply latest psu and other patches reported by etcc in database(we take downtime here)

2)Apply latest cpu patches and other reported patches as per etcc in middle tier(online)

3)Apply AD/TXK delta 12 patches(online )
4)Apply ATG 6 patches(online)
5)Apply JWS patches(online)

We need your advice if steps 2,3,4 and 5 can be done in single patching cycle.

Thanks,
Satish

Reply | Threaded
Open this post in threaded view
|

Re: Reduce downtime for patching

ErmanArslansOracleBlog
Administrator
Okay. So you want to avoid 2 cutovers , or maybe even more..
Having a up-to-date version AD-TXK is better. If you have the latest patching tools (this is related with having latest AD-TXK in place), you will directly reduce the risk for future patching and get rid of the things that may negative patch performance impact.. In addition to that, most of the time new patches are tested with the latest AD-TXK so it is a good thing to have latest AD-TXK in place.
My advice is to have a isolated patching cycle for AD-TXK.. Get them in place and then in another patching cycle apply the rest of the patches.. This is what my recommendation is.. So, in my opinion; 2 cutovers is better in my opinion.
Reply | Threaded
Open this post in threaded view
|

Re: Reduce downtime for patching

satish
Dear erman,

Thanks for your advice.

How to handle multiple restore points in this case.During every step, we create restore point..When system is released to end users and if some issue is reported..can we use restore point created at step 1 to revert the database but how to handle file systems because multiple times file systems changes..

Any suggestions how can we better handle backups during this scenario

1)seperate patching cycle for ad/tx
2)seperate patching cycle for atg
3)seperate cycle for JWS

Thanks,
Satish
Reply | Threaded
Open this post in threaded view
|

Re: Reduce downtime for patching

ErmanArslansOracleBlog
Administrator
Handling the fs-side is difficult, a fs switch may be needed there.
See the following thread , we are discussing a similar case in the following thread ->

https://erman-arslan-s-oracle-forum.124.s1.nabble.com/Module-go-live-td8391.html